JPH10136436A - Learning data collection device for wireless mobile station position detection - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a learning data collecting device which is used for a mobile radio communication system that can efficiently collect learning data and has a position detection function of a mobile station. SOLUTION: This device which collects leaning data that is used for the position detection of a mobile radio communication system that has a position detection function of a mobile station is provided with a map displaying means 210 which shows a map of a surrounding area of measuring point, an inputting means 220 which designates a measuring point on the map, an electric field intensity measuring means 230 which measures receiving electric field intensity from plural base stations according to a measurement instruction from the means 220 and a measuring point position detecting means 240 which converts the measuring point from a measurement instruction position on the map into position information, and detects an electric field intensity value of receiving electric field intensity and the position information of a measuring point according to a measurement instruction on a displayed map. It can automate the association of the electric field intensity and the position information of a measuring point.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a mobile radio communication system having a position detecting function for detecting the position of a person carrying a mobile station of mobile radio communication or an object accompanying the mobile station. The present invention relates to a learning data collection device for collecting learning data used for detection, and more particularly to a device that enables efficient collection of learning data.

[0002]

2. Description of the Related Art In recent years, in the field of mobile radio communications such as mobile phones and PHSs, the position of a mobile station or a vehicle equipped with the mobile station is detected by detecting the position of the mobile station. Have been proposed. For example, "A study on mobile object position detection based on reception level information"
(IEICE Autumn Meeting, B-269 (199
In 3)), by utilizing the fact that the wireless zones of a plurality of base stations overlap each other, a mobile station receives a received electric field strength from a plurality of base stations and a mapping table of its position (X, Y). , A method of specifying the current position of a mobile station is presented.

In such a position detection method, learning data for position detection, that is, data indicating the relationship between the received electric field strengths received by a mobile station from a plurality of base stations and the base station ID is previously stored. It is necessary to collect in many places, and the collection method becomes an issue.

As shown in FIG. 21, the conventional electric field intensity measuring device includes an electric field intensity measuring unit 2120 for measuring electric field intensity received from a plurality of base stations, and a display unit 2130 for displaying a measurement result. In the electric field intensity measuring apparatus 2110, the electric field intensity received from each base station measured by the electric field intensity measuring section 2120 is displayed on the display section 2130 in association with the base station ID as shown in FIG.

[0005] Also, "Automatic electric field strength measurement system for mobile communication" (IEICE Spring Conference, B-352 (19)
94)), an automatic electric field strength measurement system using GPS and map data is presented. As shown in FIG. 23, the measurement system includes, as a measurement system 2310, a position detection unit 2320 for detecting a current position, an electric field intensity measurement unit 2340 for measuring electric field intensity received from a plurality of base stations, and a position detection unit. 2320 and an information processing unit 2330 for processing data obtained by the electric field strength measurement unit 2340, and an FD for recording the processed data
2331, and a position detecting section 2320 includes a direction sensor 2322 for detecting a direction and a distance sensor 23 for detecting a traveling distance.
23, a gyro 2324 holding a fixed direction, a CD-ROM 2325 storing road information, a control unit 2327 controlling each unit to detect the current position, and a display unit 2326 displaying a detection result. doing.

In this system, a position detecting section 2320 generates a running history from the position information of the direction sensor 2322, the distance sensor 2323 and the gyro 2324, and the running history and the CD-RO are generated.
The correct current position is estimated by comparing with the road information stored in M2325. For this estimation, GPS 23
Use information from 21.

[0007] The electric field strength measuring section 2340 measures the electric field strength of radio waves received from a plurality of base stations.

[0008] The information processing unit 2330 records the electric field intensity measurement information obtained by the electric field intensity measurement unit 2340 and the position information detected by the position detection unit 2320 in the FD 2331. By computer processing the data recorded in the FD together with the map data, the measurement results of the electric field strength can be automatically plotted on a map.

[0009]

However, in the conventional electric field intensity measuring apparatus shown in FIG. 21, it is necessary for a measurer to manage data such as coordinate values of measurement points required for position detection separately from the measurement data. . In addition, as a preparation before measurement, it is necessary to make a specific, detailed and accurate plan for setting measurement points. For example, a work of measuring a distance on a map and entering each measurement point is performed. Have been. Further, after the measurement, an enormous amount of work occurs, such as calculating the coordinate value of each measurement point in order to associate the measurement data with the coordinates of the measurement point.

Further, the conventional device shown in FIG. 23 is a device that uses a direction sensor, a distance sensor, and a gyro, and is premised on on-board measurement. Is not available. In addition, the use of GPS is limited to outdoors and cannot be used for indoor measurements. Also, it cannot be used in places where GPS is not available due to the influence of buildings.

In general, the unit by which a person recognizes a place is:
In many cases, the coordinates are not the coordinate values, but floors and room numbers when indoors, building names and unique area names when outdoor. However, in the conventional apparatus, position detection based on coordinate values is mainstream, and there is no idea to detect a position in association with such an area. If the position represented by discrete values such as room numbers and building names is to be obtained by position detection based on coordinate values, accurate scales are required for roads, buildings, and room layouts on a map. It is uneconomical.

Further, in the conventional apparatus, it is necessary to judge on the spot of the measurement how much density of learning data for position detection and how many times measurement should be performed at the same point. Can not. As a result, inaccurate learning is not possible, such as collecting learning data again later or collecting unnecessary data to avoid unnecessary data collection accuracy without obtaining the predetermined position detection accuracy. Data collection is taking place.

[0013] Further, it is required that the learning data always be data that matches the latest radio wave condition. If there is a significant change in the radio wave condition due to the increase or decrease in the number of base stations, new dismantling of a building, or a change in the office layout, the accuracy of position detection will be reduced unless the electric field strength is measured again.

The present invention meets the above-mentioned problem and provides a mobile station position detection function capable of efficiently collecting learning data and collecting learning data realizing high position detection accuracy. It is an object of the present invention to provide a learning data collection device used for a mobile radio communication system having the same.

[0015]

Therefore, in the learning data collecting apparatus of the present invention, a map display means for displaying a map around the measurement point, an input means for designating the measurement point on the displayed map, and an input means Electric field intensity measuring means for measuring electric field intensity received from a plurality of base stations in accordance with a measurement instruction from the apparatus, and measurement point position information detection for converting the measurement point into position information from an indicated position of the measurement point on a displayed map. Means are provided.

Further, there are provided means for displaying the measurement points in a mesh on the displayed map and displaying the measurement point positions so that the measurement points are uniformly distributed in each area.

Further, means for detecting an area, not a coordinate value, is provided in correspondence with the measurement point.

Further, there is provided means for designating a continuous position as a measurement point so that measurement is performed while moving.

Further, means is provided for confirming the mathematical distance between the learning data and the position detection accuracy during the collection of the learning data.

Further, there is provided means for detecting a change in the radio wave condition and detecting learning data to be updated.

By such means, the electric field strength value of the received electric field strength is automatically associated with the position information of the measurement point, and learning data capable of improving the position detection accuracy can be collected. it can. Further, it is possible to efficiently collect learning data that meets the required accuracy.

[0022]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The invention according to claim 1 of the present invention is directed to a learning data collecting apparatus for collecting learning data used for position detection of a mobile radio communication system having a function of detecting the position of a mobile station. Map display means for displaying a map around the measurement point, input means for designating the measurement point on the displayed map, and electric field for measuring the reception field strength from a plurality of base stations a specified number of times in accordance with a measurement instruction from the input means Intensity measuring means and measuring point position information detecting means for converting the measuring point into position information from the indicated position of the measuring point on the displayed map are provided, and the measurement instruction on the map displayed on the displaying means is provided. The electric field intensity value of the received electric field intensity and the position information of the measurement point are detected, and the association between the electric field intensity value and the position information of the measurement point is automatically performed.

According to a second aspect of the present invention, the measuring point position information detecting means is configured to convert the measuring point into a coordinate value from the designated position of the measuring point on the displayed map and the map data. Thus, the association between the electric field intensity value in the learning data and the coordinate value of the measurement point can be automated.

According to a third aspect of the present invention, there is provided a mesh measuring point designating means for displaying measuring points on the map displayed on the map displaying means as mesh grid points in accordance with the scale of the map. Thus, the measurement intervals can be kept uniform. The measurement interval of the learning data has a strong correlation with the accuracy of position detection. By keeping the measurement interval uniform, the position detection accuracy can be improved.

According to a fourth aspect of the present invention, the measurement range on the map displayed on the map display means is divided into several regions,
Area measurement point indicating means for generating and displaying measurement points so that the measurement points are equal in each area is provided, and is provided for each area such as a room, a corridor, a road, a sidewalk, and a building. Measurement points can be kept uniform. By changing the density of the measurement points for each area in this manner, it is possible to collect data efficiently in accordance with required accuracy.

According to a fifth aspect of the present invention, the measurement point position information detecting means converts the designated position of the measurement point on the displayed map into an area to which the measurement point belongs.
Measurements in a specific area can be treated as data of the same point (area). By doing so, it is sufficient to collect data evenly within the area without worrying about fine positions such as the coordinate values of the measurement points, so that the efficiency of data collection is improved.
Further, unlike the detection based on the coordinate values, the position can be detected even if the scale of the map to be used is not accurate, so that the creation and preparation of the map to be used become very easy.

According to a sixth aspect of the present invention, in a learning data collecting apparatus for collecting learning data used for position detection of a mobile radio communication system having a function of detecting a position of a mobile station, a map around a measurement point is displayed. Map display means, multi-point input means for specifying the start point, intermediate point and end point of the measurement on the displayed map, and a fixed distance from the start point to the end point input from the multi-point input means. Electric field intensity measuring means for measuring the electric field intensity received from a plurality of base stations each time it moves, and detecting the coordinates of continuous measurement points between the start point, intermediate point and end point input from the multipoint input means Learning data is continuously collected by measuring the electric field strength while moving between the starting point and the ending point. As compared with the case of performing by one data collection, even the same as the time of the amount of data collected can collect training data that can dramatically improve the accuracy of the position detection result.

According to a seventh aspect of the present invention, there is provided a data distance calculating means for calculating a mathematical distance between two collected learning data, new learning data calculated by the data distance calculating means, and existing learning data. And a distance comparison measurement point designating means for designating a new measurement point based on a mathematical distance between the learning data and the learning data. If the variance of the mathematical distance of the measured data is large, increase the number of data collection at the measurement point, or if the mathematical distance between the two points is large, collect data at the midpoint between the two points Thus, learning data can be efficiently collected, and the position detection accuracy can be improved.

According to the present invention, a position detecting means for detecting a position by using collected learning data based on electric field strengths received from a plurality of base stations, and the position detecting means transmits new learning data. Error comparison measurement point indicating means for indicating a new measurement point based on an error between a position detected as learning data for position detection and a position detected without using new learning data; Yes, the position detection accuracy is checked during learning data collection, and if the required accuracy is not obtained, the number of times of learning data collection at the same point is increased, or data collection is performed in the middle between two points.
Learning data can be efficiently collected, and the position detection accuracy can be improved.

According to a ninth aspect of the present invention, there is provided a learning data collecting apparatus for collecting learning data used for position detection of a mobile radio communication system having a function of detecting a position of a mobile station, wherein the plurality of base stations are periodically arranged. Periodic electric field intensity measuring means for measuring the received electric field intensity from the data, data distance calculating means for calculating a mathematical distance between two learning data, and new learning data calculated by the data distance calculating means and existing learning data. Based on the mathematical distance between the base station, the update learning data detecting means for detecting the base station having a large change in the radio wave state and the learning data in the range affected by the change in the radio wave state, and the update learning data detecting means detects the update. A learning data management means that deletes learning data and retains only learning data that matches the latest radio wave condition is provided. Monitoring the change in the significant wave state due to influences such as layout change of office, by keeping the latest learning data, it is possible to improve the position detection accuracy of the result.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.

(First Embodiment) A configuration of a learning data collecting apparatus according to a first embodiment of the present invention will be described together with a mobile radio communication system having a position detecting function to which this apparatus is applied.

As shown in FIG. 2, the mobile radio communication system includes a mobile station 110 and a plurality of base stations 121, 122, 123.
And a communication control unit 130 that controls communication with the plurality of base stations 121 to 123, and a learning data collection device 160 that collects learning data by measuring received electric field strengths from the plurality of base stations 121 to 123.
And manages the collected learning data, and based on the received electric field strengths from the plurality of base stations detected by the mobile station 110, the mobile station 11
A position information center 150 for detecting the position of 0, and a communication control unit 13
0 and a network 140 connecting the location information center 150. The location information center 150 includes a communication control unit 151 that controls transmission and reception through the network 140, and a learning data management unit 152 that manages learning data used for location detection. And a position detection unit 153 for detecting the position of the mobile station 110 using the data of the learning data management unit 152.

Further, as shown in FIG. 1, the learning data collection device 160 includes a map display section 21 for displaying a map around the measurement points.
0 and an input unit 220 for specifying a measurement point on the displayed map
And an electric field intensity measurement unit 230 that measures reception electric field intensity from a plurality of base stations a specified number of times according to a measurement instruction from the input unit 220.
And a measuring point coordinate detecting unit 24 for converting the measuring point into a coordinate system from the indicated position of the measuring point on the displayed map and the map data.
0, learning data management unit 2 that manages collected learning data
50.

This radio communication system roughly performs the following two operations as position detection operations.

One is to use the learning data collection device 160 to determine in advance at a plurality of measurement points in the service area.
This is a learning data collection process of collecting learning data including position information and received electric field strengths from a plurality of base stations.

The other is to store the learning data in the learning data management unit 152 of the location information center 150 and to use the stored learning data to calculate the received electric field strength from a plurality of base stations measured by the mobile station. This is a position detection process for estimating the position of the mobile station.

Several methods have been proposed for position detection in this position detection processing. The present inventors also use the learning data to learn the correlation between the received radio field intensity of a plurality of base stations and the measurement position using a neural network, and the mobile station can detect the received radio field intensity from a plurality of base stations at an arbitrary point. Japanese Patent Application No. 8-212187 proposes a position detection method for estimating a position at which a measurement result is obtained based on a correlation obtained by learning. Here, the method of position detection will not be described in detail.

FIG. 3 is a flowchart of an operation performed by the learning data collection device 160 until the learning data collection device 160 collects learning data including position information at a plurality of measurement points in a service area and received electric field strengths from a plurality of base stations. It will be described using FIG.

Step 310: learning data collection device 200
Displays a map around the measurement point on the map display unit 210.

Step 320: The user moves to the measurement point while viewing the displayed map, and designates the measurement point on the displayed map from the input unit 220.

Step 330: The electric field strength measuring section 230
According to the measurement instruction from the input unit 220, the received electric field strengths from a plurality of base stations are measured a specified number of times.

Step 340: Measurement point coordinate detection section 240
Converts the measurement point into a coordinate value from the indicated position of the measurement point on the displayed map and the map data.

Step 350: learning data management section 250
Associates a pair of received electric field strengths from a plurality of base stations and a base station ID with a coordinate value of a measurement point, and manages it as learning data.

Step 360: The user repeats this step while moving the measurement point, and ends the measurement.

Here, the instruction of the measurement point on the displayed map at step 320 will be described with reference to a display example of FIG.

In this example, a map of one indoor floor is used for explanation. As shown in FIG.
Will be displayed. For example, when measuring room a,
The user moves to the measurement point in the room a, and when the user arrives at the measurement point, clicks the measurement point on the displayed map. With this operation, the measurement at the measurement point is started, and step 330
~ Learning data is generated by the procedure of step 350,
The generated learning data is managed under the learning data management unit 250. At this time, it is left to the user in which room to measure and at which point in the room to measure.

Here, a floor plan is used as an example of indoors, but any map or image data such as a road map or a house map can be used for a detailed layout diagram or outdoors.

In this embodiment, the learning data management unit 250 is provided in the learning data collection device 200. However, the learning data management unit 250 is provided in the position information center 150, and every time learning data is collected. Learning data collection device for learning data
It is also possible to configure to transmit from 200 to the location information center 150.

(Second Embodiment) The learning data collecting apparatus of the second embodiment can set measurement points at regular intervals.

As shown in FIG. 5, the learning data collection device automatically generates and displays measurement points as mesh grid points on the map displayed on the map display section 510 according to the scale of the map. A mesh measurement point indicating unit 560 is provided. Other configurations are the same as those of the first embodiment (FIG. 1).

In this apparatus, learning data is collected according to the procedure shown in the flowchart of FIG.

Step 610: Learning Data Collection Device 500
Displays a map around the measurement point on the map display unit 510.

Step 615: Mesh measurement point indicating section 56
0 automatically generates and displays measurement points as mesh-like grid points on the map displayed on the map display unit 510 according to the scale of the map. For example, if the measurement interval of learning data is 5 m
Then, the display is adjusted to the scale of the map so that the interval between the grid points is 5 m, and the grid points of the mesh shape are displayed. FIG. 7 shows a display example at this time.

Step 620: The user moves to that point using the grid point on the displayed map as a guide of the measurement point, and instructs the measurement point on the displayed map from the input unit 520.

The subsequent operation is the same as that of the first embodiment, and the electric field strength measuring section 530 measures the received electric field strengths from a plurality of base stations a specified number of times in accordance with the measurement instruction from the input section 520 (step). 630). The measurement point coordinate detection unit 540
The measurement point is converted into a coordinate value from the indicated position of the measurement point on the displayed map and the map data (step 640).
The learning data management unit 550 manages pairs of the received electric field strengths from the plurality of base stations and the base station IDs as learning data in association with the coordinate values of the measurement points (step 650). The user repeats this step while moving the measurement point, and ends the measurement (step 660).

The greatest difference between the second embodiment and the first embodiment is that the intervals between the measurement points are maintained at equal intervals by displaying the mesh-like grid points as a guide for the measurement points of the learning data. It is a point that becomes possible. FIG. 8 illustrates the relationship between the measurement interval of the learning data and the position detection error. As shown in this figure, there is a strong correlation between the accuracy of position detection and the measurement interval of learning data,
Accuracy can be improved by keeping the measurement intervals uniform using the method described in the second embodiment.

Note that the measurement point is not necessarily on the grid point, but the grid point may be used as a guide and another point may be set as the measurement point. Thereby, even in the case where a grid point on the map cannot be measured, for example, in the case of a wall at the boundary between a corridor and a room, the user skips the measurement point or measures at a nearby measurable point. It is possible.

(Third Embodiment) The learning data collecting apparatus according to the third embodiment can set measurement points at regular intervals according to each area.

As shown in FIG. 9, the learning data collection device divides the measurement range on the map displayed on the map display section 910 into several regions, and the measurement points become equal in each region. Thus, an area measurement point indicating section 960 for automatically generating and displaying measurement points is provided. Other configurations are the same as those of the first embodiment (FIG. 1).

The operation until the learning data collecting apparatus collects learning data consisting of position information and received electric field strengths from a plurality of base stations at a plurality of measurement points in the service area is the same as that of the second embodiment. Is the same.

The biggest difference between the second embodiment and the third embodiment is that in the second embodiment, mesh-like grid points are equally divided as measurement points without considering the structure in the area. On the other hand, in the third embodiment, rooms, corridors, roads,
The point is that the measurement points are set so as to keep the measurement points uniform for each area such as a sidewalk or inside a building.

FIG. 10 is a display example of the measurement points by the area measurement point indicating unit 960. There are various methods for equally providing the measurement points in the area, but here, a description will be given using a mesh-like lattice point as an example.

In a rectangular area such as a room, if the measurement interval of the learning data is 5 m, the display is adjusted to the scale of the map so that the interval between the grid points is 5 m, and the mesh grid points are displayed. At this time, the grid points are arranged so as to be as uniform as possible in the area. Areas with a small width, such as corridors and roads, are considered as a single line, and are 5m along the center line.
Set the interval measurement points.

The measurement interval may be the same 5 m in all regions, or may be changed for each region. In the example of FIG. 10, the measurement interval is changed between the room d and the other rooms a, b, c, and e. As a result, dense measurement points can be set in places where the accuracy of position detection is desired to be increased, and coarse measurement points can be set in places where accuracy is not high. For example, a living room can be measured densely to improve accuracy, but a toilet or a corridor can be roughly measured.

(Fourth Embodiment) The learning data collecting apparatus according to the fourth embodiment collects learning data as data in units of areas.

As shown in FIG. 11, the learning data collecting apparatus includes a measurement point area detection unit 1140 for detecting a measurement point not in a coordinate value but in an area. Other configurations are the same as those of the first embodiment (FIG. 1).

The operation of the learning data collection device is shown in the flowchart of FIG. In this procedure, the operation up to Step 1230 of measuring the received electric field strengths from a plurality of base stations at the measurement points in the service area is the same as Step 3 in the flowchart (FIG. 3) of the first embodiment.
This is the same as the operation up to 30.

The biggest difference between the first embodiment and the fourth embodiment is that in the first embodiment, the measuring point coordinate detection unit 240
Converts the measurement point into coordinate values (step 340), whereas in the fourth embodiment, the measurement point area detection unit 1140
Is the point to which the measurement point belongs, and the measurement point is indicated by the region name, the region ID, or the like (step 1240).

Then, learning data management section 1350 manages the pair of the received electric field strength from a plurality of base stations and the base station ID as learning data in association with the area, not the coordinate value of the measurement point (step 1250). ). That is, the measurement in the specific area is treated as data of the same point (area).

For example, the learning data is as follows:
All the measurement results in the room A are treated as an area name of the room A.

Learning data = ((CS-ID ₁ , electric field strength 1), (CS-ID ₂ , electric field strength 2),... (CS-I
D _i , electric field strength i), (area name of measurement point)) Therefore, regarding data collection in the area, data collection may be performed evenly in the area without worrying about fine positions such as coordinate values. Therefore, the efficiency of data collection is improved.

In order to convert a measurement point into a coordinate value, an accurate scale is required. In this embodiment, however, it is only necessary to know in which area the measurement point falls.
The scale of the map to be used does not need to be accurate, and the creation and preparation of the map to be used become very easy.

(Fifth Embodiment) The learning data collecting apparatus according to the fifth embodiment performs measurement while continuously moving the measurement position.

As shown in FIG. 13, the learning data collecting apparatus has a map display unit 1310 for displaying a map around a measurement point.
And a multi-point input unit 1320 for specifying a measurement start point, a plurality of intermediate points, and an end point on the displayed map, and reception electric field strengths from a plurality of base stations periodically at predetermined fixed intervals. , A continuous measurement point coordinate detection unit 1340 for detecting the coordinates of a continuous measurement point between the specified measurement start point and the end point, and learning for managing the collected learning data. And a data management unit 1350.

FIG. 14 is a flowchart showing the operation of the learning data collection device when collecting learning data at a plurality of measurement points in the service area.

Step 1410: Learning Data Collection Device 13
00 displays a map around the measurement point on the map display unit 1310.

Step 1420: The user moves to the point where measurement is started while viewing the displayed map, and inputs the measurement start point P on the displayed map from the multipoint input unit 1320.
Indicate 0. Field strength measuring section 1330 measures the received field strength from a plurality of base stations at the starting point.

Step 1430: The user then moves to the point to be measured. The electric field strength measurement unit 1330 periodically repeats the measurement of the electric field strength received from a plurality of base stations at regular intervals while the user is moving.

Step 1440: During this movement, when the moving speed is greatly changed, or when the user makes a non-linear movement such as turning a corner, a multi-point input unit is used.
From 1320, the point is displayed on the displayed map at the midpoint Pi.
(I = 1, 2,...).

Step 1445: Electric field strength measuring section 1330
Measures the received electric field strength from a plurality of base stations at the point when the intermediate point is designated.

Step 1450: When reaching the point where the measurement is to be completed while moving the measurement point, the user inputs the end point Pe on the displayed map from the multipoint input unit 1320.

Step 1455: When an end point is specified, electric field strength measuring section 1330 measures the received electric field strength from a plurality of base stations at the end point.

Step 1460: Continuous measuring point coordinate detecting section
1340 detects the coordinate value of each measurement point.

At this time, the continuous measuring point coordinate detecting unit 1340
The coordinate value of each measurement point is detected as follows.

The start point, the intermediate point, and the end point are indicated by coordinates of the start point P0, the intermediate point Pi (i = 1, 2,...), The end point Pe on the displayed map, and the map data. Convert to a value.

The points between the intermediate points are obtained as follows.

The number of times of measurement between the two intermediate points Pi-1 and Pi by the electric field strength measuring section 1330 is defined as Ni. However, Ni does not include the measurement at the intermediate point Pi-1, but includes the measurement at the intermediate point Pi.

Actual measurement point q between two points Pi-1 and Pi
ij (j = 1, 2,..., Ni) can be approximated by qij = Pi-1 + (Pi-Pi-1) * j / Ni. Note that q00 = P0.

Taking two-dimensional coordinates as an example, Pi = (Xi, Y
i) (i = 0, 1,..., e), qij = (Xi-1 +
(Xi-Xi-1) * j / Ni, Yi-1 + (Yi-Yi-1) * j
/ Ni) (i = 0, 1,..., E) (j = 1, 2,.
i).

The total number of measurements from the start point to the end point is 1 + ΣNi (i = 1,..., E).

Step 1470: Learning data management section 135
0 manages the pair of the received electric field strength from a plurality of base stations and the base station ID as learning data in association with the coordinate value of the measurement point detected by the continuous measurement point coordinate detection unit 1340.

Step 1480: The user repeats this step while moving the measurement point, and ends the measurement.

As described above, this learning data collecting apparatus
Collect data while constantly changing measurement points. Compared to the method of stopping at the same point and repeating the measurement multiple times, this collection method can collect the learning data from a wide range of points, although the degree of labor is the same. The effect can be enhanced.

That is, when looking at two measurement points,
Compared to the method of collecting data N times only at two measurement points, in this method of continuously collecting data 2N times between two measurement points, the data amount and the labor of collection are the same, The accuracy of the position detection result is dramatically improved.

(Sixth Embodiment) The learning data collecting apparatus according to the sixth embodiment adjusts the number of measurements and the measurement position while observing the measurement data.

As shown in FIG. 15, the learning data collecting apparatus calculates a mathematical distance between two learning data, and calculates a mathematical distance between the new learning data and the existing learning data. And a distance comparison measurement point indicating unit 1570 for indicating a new measurement point for comparison. Other configurations are the same as those of the first embodiment (FIG. 1).

The operation of this apparatus will be described with reference to the flowchart of FIG.

Step 1610: Learning Data Collection Device 15
00 displays a map around the measurement point on the map display unit 1510.

Step 1620: The user moves to the measurement point while viewing the displayed map, and designates the measurement point on the displayed map from the input unit 1520.

Step 1630: Electric field strength measuring section 1530
According to the measurement instruction from the input unit 1520, measures the received field strengths from a plurality of base stations a specified number of times.

Step 1640: Measurement point coordinate detection section 1540
Converts the measurement point into a coordinate value from the indicated position of the measurement point on the displayed map and the map data.

Step 1650: Learning data management section 1550
Manages pairs of received field strengths from a plurality of base stations and base station IDs as learning data in association with coordinate values of measurement points.

The operation up to this point is the same as the operation in the first embodiment.

Step 1660: Data distance calculation section 1560
Is the mathematical distance between the new training data measured and the existing training data measured at the same measurement point, and the mathematical distance between the new training data and the existing training data around the measurement point Is calculated.

Here, a method of calculating the mathematical distance will be described. The mathematical distance is calculated using an arbitrary distance function ρ. The distance function ρ has the following condition, ie, a non-negative real number ρ (x, y) uniquely corresponds to any binary x, y of the set X: 1) ρ (x, x) = 0 , Conversely, if ρ (x, y) = 0, then x
= Y 2) ρ (x, y) = ρ (y, x) 3) For any three points x, y, z, ρ (x, z) ≦ ρ (x, y) + ρ (y, z) ) Is a function that satisfies the condition

For example, when the Euclidean distance, which is a general concept of distance, is used, the mathematical distance is calculated as follows.

Among the base stations with the strong electric field strength, the top four base stations are B1, B2, B3, B4, and B1, B2, B
It is assumed that E1, E2, E3, and E4 electric field strengths have been received from each of B3 and B4. The data distance calculation unit 1560 calculates the electric field strength data ((B1, E1), (B2, E2),
(B3, E3), (B4, E4)) and the j-th field strength data ((B
Sj1, Ej1), (BSj2, Ej2), (BSj3, Ej3),
(BSj4, Ej4)) (however, BSj1, BSj2, BSj3,
BSj4 is the top four base stations among the base stations with strong electric field strength received at the j-th measurement point (Xj, Yj),
Ej1, Ej2, Ej3, and Ej4 indicate the electric field strengths received from these base stations, respectively).
In the case of B2 = BSj2, B3 = BSj3, B4 = BSj4 ^{is, ((Ej1-E1) 2} + (Ej2-E2) 2 + (Ej3-E3)
² + (Ej4-E4) ² ) Calculate with ^1/2 .

If there is a base station that does not match among the base stations to be compared, the mathematical distance is calculated by setting the electric field strength of the base station that does not match to 0.

In this way, the data distance calculation unit 1560
After calculating the mathematical distance between the new learning data and the existing learning data, Step 1670: The distance comparison measurement point indicating unit 1570 checks whether or not this mathematical distance is larger than a predetermined value. If the variance of the mathematical distance of the measurement data at the same point is large, Step 1680: Distance comparison measurement point designating section 1570 instructs input section 1520 to increase the number of data collections at the measurement point. If, in step 1670, the mathematical distance between the two points is large, in step 1680,
It instructs to collect data at the midpoint between these two points.

Step 1690: The user repeats this step while moving the measurement point, and ends the measurement.

As described above, this learning data collecting apparatus
By calculating the data distance of the learning data during the learning data collection, if the variance of the mathematical distance of the measurement data at the same point is large, the number of data collection at the measurement point is increased and the mathematical distance between the two points is increased. When the distance is large, by collecting data at an intermediate point between the two points, learning data can be efficiently collected, and the position detection accuracy can be improved.

The data distance calculation section and the distance comparison measurement point designating section in this embodiment are not limited to the apparatuses of the second, third, fourth, and fifth embodiments as well as the first embodiment. Can be applied in a similar manner.

(Seventh Embodiment) The learning data collecting apparatus of the seventh embodiment adjusts the number of learning data to be collected so as to obtain a predetermined position detection accuracy.

As shown in FIG. 17, the learning data collecting apparatus includes a position detecting unit 1760 for calculating a current position using learning data based on the received electric field strengths from a plurality of base stations, and a new learning data. An error comparison measurement point instructing unit 1770 is provided for examining an error between a case where it is used as learning data for position detection and a case where it is not used, and instructing a new measurement when the error is large. Other configurations are the first embodiment (FIG. 1)
There is no change.

The operation of the learning data collection device is shown in the flowchart of FIG. Step 1 in this procedure
The operation from step 810 to step 1850, that is, a map around the measurement point is displayed on the map display unit 1710, and the reception electric field strengths from a plurality of base stations are measured a specified number of times at the measurement point indicated on the map. Is converted into coordinate values, and the operation of managing the learning data in which the measurement data and the coordinate values are associated with each other by the learning data management unit 1750 is the same as in the first embodiment.

Step 1860: The position detecting section 1760 adds the newly measured learning data to the learning data, calculates the current position based on the measured electric field strengths received from the plurality of base stations, and calculates a new position. The current position is calculated based on the received electric field intensity without adding the measured learning data to the learning data. The error comparison measurement point indicating unit 1770 is
60 calculates the error between the current position calculated by adding the new learning data to the position detection learning data and the current position calculated without adding the new learning data to the position detection learning data. Step 1870: If the error is greater than the required accuracy, step 1880: instruct input unit 1720 to increase the number of times of learning data collection at the same point or to collect data around the measurement point.

Step 1890: The user repeats this step while moving the measurement point, and ends the measurement.

As described above, the apparatus of this embodiment checks the position detection accuracy during learning data collection, and if the required accuracy cannot be ensured, increases the number of times learning data is collected at the same point, The number of learning data is increased by, for example, collecting data in the middle between the two points. By doing so, it is possible to efficiently collect learning data for obtaining a predetermined position detection accuracy.

The position detecting section and the error comparison measuring point indicating section in this embodiment are not limited to the apparatuses of the second, third, fourth, fifth and sixth embodiments as well as the first embodiment. The same can be applied to this.

(Eighth Embodiment) The learning data collecting apparatus according to the eighth embodiment updates the learning data according to a change in the radio wave condition.

As shown in FIG. 19, the learning data collecting apparatus calculates a mathematical distance between two learning data and a periodic electric field intensity measuring unit 1910 which periodically measures electric field intensity received from a plurality of base stations. A data distance calculation unit 1920, and an update learning data detection unit 1930 that detects a base station having a large change and a range affected by the base station and learning data to be updated when there is a significant change in radio wave state. And a learning data management unit 1940 for adding and deleting learning data.

The operation of the learning data collection device will be described with reference to the flowchart of FIG. Step 2010: The periodic electric field strength measuring unit 1910 periodically measures electric field strengths received from a plurality of base stations while the position detection system is operating.

Step 2020: Data distance calculation section 1920
Calculates the mathematical distance between the measured new training data and the existing training data measured at the same measurement point.

Step 2030: As a result, if the mathematical distance is larger than a predetermined value, Step 2040: The update learning data detecting unit 1930 detects a base station having a large change, and detects an influence of the base station. The range and the learning data to be updated are detected.

The detection of a base station having a large change is performed when the electric field strength from the base station Bi is Bio for old learning data and Bin for new learning data. | Bin−Bio | (| X |
Is a function for calculating an absolute value). The range affected by the base station Bi is all the learning data including the electric field strength of the base station Bi, and these are the learning data to be updated.

Step 2050: Learning data management section 1940
Is used for the learning data to be updated
Delete old learning data and add new learning data.

As a result, a large change in the radio wave condition due to the influence of the increase or decrease of the base stations, the new dismantling of the building, the change of the office layout, etc. is monitored, the old learning data affected by the change is deleted, and the latest learning By keeping the data,
The accuracy of the position detection result can be improved.

Note that the data distance calculation unit 1920, the updated learning data detection unit 1930, and the learning data management unit 1940 in this embodiment are not provided in the learning data collection device 1900,
It can also be placed in the location information center 150 of FIG. In this case, the learning data collection device 1900 and the location information center 150
The data is communicated between the two, and a mathematical distance between the learning data is calculated in the position information center 150, and the learning data is updated.

Also, by installing the learning data collection device 1900 in the base stations 121, 122 and 123 in FIG. 2, the data collection device for periodically measuring the electric field intensity received from a plurality of base stations is installed. Without creating a new location,
It is possible to monitor a large change in radio wave condition. Thus, for example, a learning data collection device with a built-in base station is used for monitoring changes periodically, and if a change is found and the change is large, learning data around the change is newly collected. This makes it possible to efficiently keep the latest learning data.

Further, the apparatus according to the first to seventh embodiments is provided with an updated learning data detecting section 1930 and a learning data managing section 1940 having a data updating function, so that the learning data collecting section holds the latest learning data. The device can also be configured. Since this device does not have the periodic electric field strength measurement unit 1910 and the data distance calculation unit 1920, it monitors large changes in radio wave conditions due to the effects of increases and decreases in base stations, new dismantling of buildings, layout changes in offices, etc. Although it does not have a function, knowing a significant change in the radio wave state by some method, the update learning data detection unit 1930 deletes the learning data in the range affected by it and collects the learning data within the range again Thus, the latest learning data can be held.

[0132]

As is apparent from the above description, the learning data collection device of the present invention can automatically calculate the coordinate values of the measurement points and automatically associate the measurement points with the measurement data.

Also, the measurement intervals are kept uniform, or
By changing the density of measurement points in each area, such as rooms, corridors, roads, sidewalks, and buildings, and maintaining uniform measurement intervals within the area, the position detection accuracy is improved, and the efficiency that meets the required accuracy is improved. Data collection becomes possible.

Further, by treating the measurement in a specific area as data of the same point (area), it is not necessary to worry about a fine position such as a coordinate value of the measurement point, and the efficiency of data collection is improved. In this case, since the position can be detected even if the scale of the map is rough, the creation and preparation of the map to be used become very easy.

In an apparatus for continuously collecting data between two measurement points, the accuracy of position detection can be dramatically improved without increasing the amount of data and the labor for collecting data.

Further, a device that calculates the data distance of the learning data and checks the position detection accuracy during the learning data collection can efficiently collect the learning data that provides a predetermined position detection accuracy.

A device for monitoring a change in radio wave condition due to an increase or decrease in base stations, a new building dismantlement, a change in office layout, or the like, maintains the latest learning data.
The accuracy of the position detection result can be improved.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of a learning data collection device according to a first embodiment;

FIG. 2 is a block diagram showing a configuration of a mobile radio communication system to which the learning data collection device according to the first embodiment is applied;

FIG. 3 is a flowchart illustrating an operation of the learning data collection device according to the first embodiment;

FIG. 4 is an example of a map display in the learning data collection device according to the first embodiment;

FIG. 5 is a block diagram illustrating a configuration of a learning data collection device according to a second embodiment;

FIG. 6 is a flowchart showing the operation of the learning data collection device according to the second embodiment;

FIG. 7 is a map display example in the learning data collection device according to the second embodiment;

FIG. 8 is a diagram showing a correlation between a learning data collection interval and a position detection result error;

FIG. 9 is a block diagram illustrating a configuration of a learning data collection device according to a third embodiment;

FIG. 10 is a map display example in the learning data collection device according to the third embodiment,

FIG. 11 is a block diagram illustrating a configuration of a learning data collection device according to a fourth embodiment;

FIG. 12 is a flowchart illustrating an operation of the learning data collection device according to the fourth embodiment;

FIG. 13 is a block diagram illustrating a configuration of a learning data collection device according to a fifth embodiment.

FIG. 14 is a flowchart showing the operation of the learning data collection device according to the fifth embodiment;

FIG. 15 is a block diagram illustrating a configuration of a learning data collection device according to a sixth embodiment.

FIG. 16 is a flowchart showing the operation of the learning data collection device according to the sixth embodiment;

FIG. 17 is a block diagram illustrating a configuration of a learning data collection device according to a seventh embodiment;

FIG. 18 is a flowchart showing the operation of the learning data collection device according to the seventh embodiment;

FIG. 19 is a block diagram illustrating a configuration of a learning data collection device according to an eighth embodiment.

FIG. 20 is a flowchart showing the operation of the learning data collection device according to the eighth embodiment;

FIG. 21 is a block diagram showing a configuration of a conventional electric field intensity measuring device.

FIG. 22 is a display example of a conventional electric field intensity measuring device,

FIG. 23 is a block diagram showing a configuration of a conventional automatic electric field intensity measuring apparatus for mobile communication.

[Explanation of symbols]

110 Mobile station 121, 122, 123 Base station 130 Communication control unit 140 Network 150 Location information center 151 Communication control unit 152, 250, 550, 950, 1150, 1350, 1550, 1750, 1940
Learning data management unit 153, 1760, 2320 Position detection unit 160, 200, 500, 900, 1100, 1300, 1500, 1700, 1900
Learning data collection device 210, 510, 910, 1110, 1310, 1510, 1710 Map display unit 220, 520, 920, 1120, 1520, 1720 Input unit 230, 530, 930, 1130, 1330, 1530, 1730, 2110, 212
0, 2340 Electric field strength measurement section 240, 540, 940, 1540, 1740 Measurement point coordinate detection section 560 Mesh measurement point indication section 960 Area measurement point indication section 1140 Measurement point area detection section 1320 Multipoint input section 1340 Continuous measurement point detection Unit 1501, 1701 Learning data collection device 1560, 1920 Data distance calculation unit 1570 Distance comparison measurement point indicating unit 1770 Error comparison measurement point indicating unit 1910 Periodic electric field strength measuring unit 1930 Update learning data detecting unit 2130, 2326 Display unit 2310 Measurement system 2321 GPS 2322 Orientation sensor 2323 Distance sensor 2324 Gyro 2325 CD-ROM 2327 Control unit 2330 Information processing unit 2331 FD

Claims (9)

[Claims] 1. A learning data collection device for collecting learning data used for position detection of a mobile radio communication system having a function of detecting a position of a mobile station, comprising: map display means for displaying a map around a measurement point; An input means for designating a measurement point on the displayed map; an electric field intensity measuring means for measuring reception electric field strengths from a plurality of base stations a specified number of times in accordance with a measurement instruction from the input means; Measuring point position information detecting means for converting the measuring point from the designated position of the measuring point to position information, and a measuring instruction on a map displayed on the display means, the electric field intensity value of the received electric field intensity and A learning data collection device for detecting the position information of the measurement point. 2. The method according to claim 1, wherein the measurement point position information detecting means converts the measurement point into a coordinate value based on a designated position of the measurement point on a displayed map and map data.
A learning data collection device according to item 1. 3. The apparatus according to claim 1, further comprising: a mesh measurement point indicating means for displaying the measurement points on the map displayed on the map display means in accordance with the scale of the map as grid points of a mesh. 3. The learning data collection device according to 2. 4. An area measurement in which a measurement range on a map displayed on the map display means is divided into several areas, and measurement points are generated and displayed so that the measurement points are equal in each area. 2. The apparatus according to claim 1, further comprising point indicating means.
Or the learning data collection device according to 2. 5. The learning according to claim 1, wherein the measurement point position information detecting means converts a designated position of the measurement point on a displayed map into an area to which the measurement point belongs. Data collection device. 6. A learning data collecting apparatus for collecting learning data used for position detection of a mobile radio communication system having a function of detecting a position of a mobile station, comprising: map display means for displaying a map around a measurement point; Multi-point input means for designating a start point, an intermediate point, and an end point of measurement on the displayed map; each time a fixed distance is traveled between the start point and the end point input from the multi-point input means. Electric field intensity measuring means for measuring electric field intensity received from a plurality of base stations; and continuous measurement for detecting coordinates of a continuous measuring point between a start point, an intermediate point and an end point inputted from the multipoint input means. A learning data collecting apparatus comprising: point coordinate detecting means; and continuously collecting learning data by measuring an electric field strength while moving between a start point and an end point. 7. A data distance calculating means for calculating a mathematical distance between two collected learning data, and a mathematical distance between new learning data calculated by the data distance calculating means and existing learning data. The learning data collection device according to claim 1, further comprising a distance comparison measurement point designating unit that designates a new measurement point based on the distance. 8. A position detecting means for detecting a position using collected learning data based on received electric field strengths from a plurality of base stations, and said position detecting means uses new learning data as learning data for position detection. An error comparison measurement point indicating means for indicating a new measurement point based on an error between a position detected by using and a position detected without using new learning data is provided. 7. The learning data collection device according to 6. 9. A learning data collection device for collecting learning data used for position detection of a mobile radio communication system having a function of detecting the position of a mobile station, comprising: A periodic electric field strength measuring means for measuring the distance between two learning data; a data distance calculating means for calculating a mathematical distance between two learning data; and a mathematical formula between the new learning data calculated by the data distance calculating means and the existing learning data. Based on the distance, update learning data detecting means for detecting a base station having a large change in the radio wave state and learning data in a range affected by the change in the radio wave state, and learning data detected by the update learning data detecting means. A learning data collection device comprising: learning data management means for deleting and holding only learning data matching the latest radio wave condition.